PHYSICAL PROPERTIES OF METALLIC MAGNETIC COMPOUNDS : 5f-LOCALIZED APPROACHThe structure, magnetic and related properties of the semimetallic, semiconducting and ionic compounds of actinides

(1)

HAL Id: jpa-00218812

https://hal.archives-ouvertes.fr/jpa-00218812

Submitted on 1 Jan 1979

HAL

is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire

HAL, est

destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

PHYSICAL PROPERTIES OF METALLIC MAGNETIC COMPOUNDS : 5f-LOCALIZED APPROACHThe structure, magnetic and related properties of the semimetallic, semiconducting and ionic

compounds of actinides

W. Suski

To cite this version:

W. Suski. PHYSICAL PROPERTIES OF METALLIC MAGNETIC COMPOUNDS : 5f-

LOCALIZED APPROACHThe structure, magnetic and related properties of the semimetallic, semi-

conducting and ionic compounds of actinides. Journal de Physique Colloques, 1979, 40 (C4), pp.C4-

43-C4-48. �10.1051/jphyscol:1979416�. �jpa-00218812�

(2)

PHYSICAL PROPERTIES OF METALLIC MAGNETIC COMPOUNDS : 5f- LOCALIZED APPROACH

The structure, magnetic and related properties of the semimetallic, semicon- ducting and ionic compounds of actinides

W. Suski

Institute for Low Temperature and Structure Research, Polish Academy of Sciences, P.O. Box 937, 50-950 Wroclaw, Poland

R6sumC. - Dans cette revue nous discutons la structure, les proprietks magnetiques et celles likes au magnetisme des composCs semim6talliques, semiconducteurs ou ioniques des actinides qui ont kt6 etudikes pendant Ies deux dernibres annkes. Nous insistons tout particulibrement sur les systbmes cubiques de structure du type Th3P4. Jusqu'h prksent, ce groupe de composes a reGu moins d'attention que par exemple les composes du type NaCl, h cause de leur degre plus grand de complexitk. On a aussi present6 les resultats obtenus pour les systbmes ternaires d'uranium de structure tktragonale, ainsi que pour quelques composks ioniques d'uranium (IV) et (v)

.

On a fait enfin une courte revue des recherches recentes sur la diffusion inelastique des neutrons.

Abstract. - In the present review we discuss the structure, magnetic and related properties of those semimetallic, semiconducting and ionic compounds of actinides which have been investigated during the last two years. The special interest is devoted to the cubic systems with Th,P, type of crystal structure. Up to now this group of compounds has been given considerably less attention than for example the NaC1-type compounds because of their higher degree of complexity. Also the results obtained for the tetragonal uranium ternaries as well as some ionic compounds of uranium (IV) and (V) are presented. Recent inelastic neutron scattering experiments are reviewed.

1. Introduction. -During the last decade we have been facing the outburst of the research on the actinide systems. The aim of this paper is to present some results which have been obtained since the 2nd International Conference on the Electronic Structure of Actinides in WrocJaw, on the semimetallic, semiconducting and ionic compounds, which are supposed to possess the localized 5f electrons. In that time the considerable effort has been made to understand the electronic structure and properties of the cubic, the NaC1- and CaF,-type compounds.

Also at this Conference a lot of papers concerning the monocompounds and dioxides will be presented, thus we are not going t o compete and these classes of compounds will be omitted in the present review.

In fact it is to be pointed out that in spite of the above mentioned effort the understanding of these simple families (as far as the crystal structure is concerned) is still far from being complete. Therefo- re one cannot wonder that the understanding of more complex compounds which will be discussed now is even less satisfactory. For this reason, first of all we should like to present the results of magnetic and related measurements, providing only very few theoretical results which are available in current literature. Although we have stated that we will not consider the monocompounds, for sake of order we should mention that recently the magnetic properties and structure of the UX-LnX solid solutions (where X, pnictide and Ln, light rare e&h element) have been extensively investigated by TroC et al. [I]. The preparation of these alloys seems to be

an important achievement because even recently some people deny the miscibility of the actinide and lanthanide monocompounds.

2. Th,P,-type compounds.

-

The structure and magnetic properties of these compounds have been investigated for several years. However, only the uranium pnictides are described in more detailed way and their single crystals are available which provide possibility of more sophisticated experiments. Some preliminary data on the crystalloche- mistry of these compounds were collected in [2]. It follows from this work and from the more recent results that at present almost exclusively only the transuranium chalcogenides of this type of structure are known. Moreover, all the chalcogenides exhibit the homogeneity range with the terminal composi- tions An3Y4-An,Y,( y ). The change of composition over the range of homogeneity is realized by crea- tion of the vacancy on the metal side [3]. As noted by Carter et al. [4] the assurnption,of mere statistical distribution of vacancies is somewhat more extreme for the pure cubic actinide sesquichalcogenides (yAn,Y,) for which Zachariasen [3] originally proposed that 4/3 vacancies are randomly distributed among the 12 metal sites. Since each metal atom has eight near metal neighbours in this structure, the probability of the pairing of vacancies is sufficiently high (8/81) and cannot be ignored even in a zero- order discussion. In figure 1 are presented the lattice parameters of the An,Y4 and y An,Y, type chalcogenides in dependence on the atomic number of actini-

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1979416

(3)

W. SUSKI

j

tellurides

4

sy)

^IA1

Fig. 1 . -The lattice parameters of the AnZ, and y An2Y3-type chalcogenides in dependence on the atomic number of actinide.

de. It follows from this figure 1 that there is no regularity in presented plots. At present we are not able to give any explanation for such behaviour. It should be mentioned that the situation for more extensive1 y examined monocompounds , specially for the lighter actinides is similar. In individual cases the difference of formal and real composition can influence the plot, the more so that the precise composition determination is sometimes difficult.

However, at least the uranium tellurides ([5] and table I) do not demonstrate any appreciable lattice parameter change over the homogeneity range. The same property is exhibited by numerous rare earth chalcogenides, but Allbutt [6] found in the Pu-S system a systematic variation of the lattice constant with composition that ranges from Pus,.,, to Pus,,,,.

The attempts to obtain the thorium chalcogenides and uranium sulphide were futile. On the contrary the thorium pnictides have been known for years to exist. The uranium pnictides as mentioned above are available in the single crystal form and some of their properties are well known. Among the transuranium Table I.

-

Structural and magnetic data of the U,Te,-, system [S].

pnictides with the Th,P, structure the antimonides of Np and Pu are known only 171.

The magnetic properties of the uranium pnictides have been reviewed several times (see eg. [8, 93).

These compounds appeared to be ferromagnetic below about 200 K, however, some questions still remain unsolved. On the basis of the crystal field considerations [lo], the three mutually perpendicu- lar magnetic sublattices below the ordering temperature can be expected to exist. This conclusion has been confirmed to some extent by both theoretical calculations (see eg. [1 11) and experiment [12].

However, none direct neutronographic proof has been provided as yet [l3]. Moreover it has been shown that if the magnitude of exchange interactions exceeds that of crystal field interactions by, say, a factor of 4, then the magnetic structure, even if essentially non-collinear, becomes asymptotically collinear [14].

In order to clarify this situation several new experiments have been done predominantly by Czech group. A study of Barkhausen effect in U3P4 single crystals confirmed that domain structure is oriented into the easy direction ( 11 1 ), and initial susceptibility exhibits high magnetocrystalline anisotropy. The average volume magnetization of which is reversed by a single Barkhausen jump is equal to 3.6 x lo-' cm3, indicating relatively large domains [l5]. Previous measurements on U,As, showed the existence of discontinuous (I order) transition in the field dependence of magnetization at H, = 16 T (18 T) at temperature 77 K (4.2 K) respectively[16]. At present KaczCr and Novotnq [17] report the results of magnetization measurement performed with the pulsed field applied in various directions in the [I101 plane. They show that the jump in magnetization can be explained by including anisotropy terms of higher order which are considerably larger than the constant K,

.

The experimentally found value of hysteresis is much smaller than the calculated one and that fact seems to be an evidence that the transition mechanism of homogeneous rotation is accompanied by some other mechanism probably domain nucleation 1171. The authors of [18] tried to support the concept of the non-collinearity of the magnetic sublattices of U3As4 on the basis of the high field susceptibility measurements and proposed the angle between the ( 11 1 ) direction and one of the sublatti- ce directions to be 82". Yet they found that the high field magnetic susceptibility is higher than in ferromagnetic metals and close to that observed for ferrimagnets with non-collinear magnetic structure.

The s-f coupling constant

r

⁼2.4 eV has been determined in the Knight shift study [l9]. The devia- tions from the Curie-Weiss law were ascribed to the conduction electron contribution to paramagnetic susceptibility. The results of law temperature heat capacity measurements [20] confirmed profound in-

(4)

THE STRUCTURE, MAGNETIC AND RELATED PROPERTIES C4-45 fluence of crystal field on the electronic structure of

U3As, and U3Sb4. The results proved that essentially only three magnetic states of uranium atom are populated. The influence of the magnetic field applied along ( 1 1 1 ) direction on the electrical resistivity of U3As4 in the vicinity of the Curie point has been investigated. Results show that the electrical resistivity is sensitive to the domain structure and at lower magnetic field the spin critical fluctuation induces a sharp maximum [21]. Moreover a broad maximum was revealed below T,. Similar type anomaly of resistivity, although less pronounced, was observed for U,P4. The results can be described by the formu- la derived by Kim [22], [23].

The uranium chalcogenides have not been so broadly investigated up to now. The selenide exhibits a maximum in the temperature dependence of magnetization at low temperatures, but the telluride was reported to be ferromagnetic (for details see 193). This last compound has been under examination recently [51. In table I some preliminary structural and magnetic data are collected. The magnetic properties of these materials are complex.

Below 100 K they undergo a transition to ordered state with a ferromagnetic contribution. However, similarly as for many other uranium ferromagnets the maximum is observed in the temperature dependence of magnetization. According to the crystal field calculation [lo], one can expect a strong single- ion-anisotropy which competing with an exchange anisotropy could be account for such a behaviour [24]. Nevertheless, the other explanation connected with a complicated domain structure seems to be more tempting [25]. The value of saturation moment which is considerably low cannot be explained satisfactory at present. At the same time it is difficult to interpret the value of the effective moment which does not exhibit any regular dependence on the composition. The observed high effective moment for U3Te4 (x = 0.33) as well as the negative Weiss constant are a puzzle. Also the temperature dependence of magnetization for this composition is different from other samples, presenting two anomalies at low temperatures. It could be that the magnetic order in U3Te4 is of the different type. Preliminary neutron diffraction experiment performed at 4.2 K has showed a weak contribution to nuclear peaks, thus one can expect a magnetic structure of ferromagnetic type only [26].

Among the chalcogenides of the transuranium elements Pu3S, is known to be antiferromagnetic below 10 K [27] while Am3Se4 and Am3Te4 do not exhibit any magnetic orderifig- down to 4.2 K [28].

The examination of the Th3X4-U3X4 type solid solutions appeared to be possible only recently after proposing a method for purification and growth of single crystals which is a modification of the Van Arkel method [29]. Th3P, appeared to be n-type semiconductor with the forbidden gap equal t o

0.43 eV [30]. The effective mass of electrons of order 0.55 no is an intermediate value for conduction and impurity band. Subsequently the examination of (Th, U,-,),As, solid solutions were carried out up to x Z= 0.83. The analysis of transport properties sho- wed that the conduction mechanism in these alloys is the same as in Th3As4. The temperature dependence of magnetic susceptibility shows a maximum at low temperatures and values of magnetic moment ran- ging from 2.7 to 2.9 BM are close to that in U3As4.

The behaviour of magnetic moment shows that 5fZ electrons of uranium do not enter the conduction band and are localized in these materials [31].

3. The ternaries. - For years the family of tetragonal compounds with the closely related structure types of : anti-Cu2Sb, PbFCl, UGeTe and U2N2Sb is examined extensively (see [8, 91). According to some theoretical predictions these compounds could be either ferromagnetic or antiferromagnetic. Re- cently the UGeY and U2N2Sb-type compounds have been investigated. The UGeY-type compounds proved to be antiferromagnetic below 88, 40 and 73 K for UGeS, UGeSe and UGeTe respectively [32]. The magnetic structure of sulphide and telluri- de corresponds to the following sequence of the ferromagnetic planes :

+ +

^{- -}^,while that of selenide :

+

^-

+

^-^[33].

The tetragonal compounds of the U2N,Z type are ferromagnetic below 166, 154 and 71 K for UzN2Sb, U2N2Bi and UZN2Te respectively [34]. U2N2Te does not reach saturation even at 14 T and a value of magnetic moment obtained at this field strength is 2.78 BM which is the highest one determined for the ferromagnetic uranium compounds. Moreover, two maxima are observed in the temperature dependence of magnetization. The reason of this unusual property appeared to be magnetic structure of U,N2Te [35]

in which the moment direction forms an angle 70 & 5" to the tetragonal axis and magnetic moment is equal to 2.50 BM. Because of the lack of saturation at 14 T one can expect that using a higher magnetic field the increase of the magnetic moment value will be observed. In fact at 33 T the value as high as 3.18 BM has been obtained, which is close to 3.3 BM, the maximum possible value for the U4' ion [36].

The other tetragonal compounds are UPd2Si, and UPd2Gez which exhibit a behaviour suggesting the existence of ferromagnetism followed by antiferro- magnetism as the temperature is increased [37].

Finally the transverse magnetoresistance of single crystal of ferromagnetic UAsSe has been measured.

It was found to be negative and anisotropic and to show a deep and finite minimum at 105.5 K which is lower than the Curie point (1 13 K) and the disconti- nuity point of the derivative of zero magnetic field resistivity (109 K) [38]. It has not been finally decided if CaU2S4 and SrU2S, are cubic or tetrago-

(5)

C4-46 W. SUSKI

nal. The preliminary magnetic susceptibility measurement suggests the presence of the U" ion in them.

At low temperature a distinctive maximum observed for both compounds in the temperature dependence of susceptibility might be due to an antiferromagnetic ordering [39].

As.the last group of semimetallic compounds the MU,Y,, type compounds (M = 3d metals or Mg, Y = S, Se and Te) will be briefly described [40].

MgU,Y,, are paramagnetic, in the case of the compounds with Co, Fe and Ni there seems to exist a magnetic ordering at low temperature but of complex type. Other compounds exhibit maxima at low temperature which might be connected with an induced antiferromagnetic ordering, although a weak ferromagnetic contribution is also observed in each of them. However, in the case of CrU,Se,, an inflection point was detected only instead of maximum.

4. Ionic compounds. - Special attention was devoted to ionic compounds of the uranium /IV/

which is placed in the antiprismatic coordination polyhedron frequently with oxygen atoms as li- gands. The crystal field potential of such polyhedron is considerably simple axial ana usually is composed of the three independent terms. The calculations gave in some special cases (USiO,, UCl,) all the crystal field parameters and in others the ground state of the uranium ion [4l]. The paramagnetic properties of U(S04),4 H,O, U,O,(OH),(SOJ, and U(OH),SO, in the terms of above assumption have been explained [42]. In the last case at 21 K the anomalies in the temperature dependence of the magnetic susceptibility and low temperature specific heat have been observed. According to the preliminary considerations these anomalies result from crystallographic transition induced by the cooperati- ve Jahn-Teller effect [43]. The influence of this effect on the paramagnetic susceptibility of S = 1 (singlet-doublet) systems has been explained by

~ o l n i e r e k [44] who discussed the thermal variation of the quadrupolar parameters and their relation to susceptibility. On the basis of the reflectance spec- tra and the temperature dependence of paramagnetic susceptibility the full scheme of crystal field levels of UP,O, (octahedral coordination) of the U4' ion has been proposed [45].

In turn the compounds of the uranium /V/ will be described. The crystal field Hamiltonian parameters and separation of the ground and the first excited levels have been determined for CsUF,, Cs,UF, and (NH,),UF, [46]. The crystal-field-ground-state of UCl, has been determined in the ESR experiment [47]. The magnetic measurements carried

out on NaUO, revealed a maximum at 32 K which is supposed to be connected to the canted spin arrangement as leading to a steep increase in the magnetic susceptibility at

-

10 K [48]. At 32 K the A type anomaly in this compound in the low temperature specific heat was observed [49].

5. Inelastic neutron scattering. - In spite of Lan- der pesimistic opinion [50] recently few papers concerning the inelastic neutron scattering have appeared. First successful experiments performed on UPd, and UAs [51] above the magnetic transition temperature gave the weak and poorely resolved lines, therefore their interpretation is ambiguous.

Successive experiment performed on UCI, proved to be more precise [52]. In table I1 [52] the energy separation A E of the

r,

^and

r,

levels in UCl, as deduced from optical, magnetic and spectroscopy data is presented. It follows from this table that the results of neutron spectroscopy are consistent with those obtained in other experiments. Recently a strong, well-resolved, neutron energy loss line by inelastic neutron scattering has been found in Table 11.

-

The energy separation of the I', and

I',

levels in UCl, as determined in various experiments.

cm-'AE meV experiment

-

- -

140,211 17.3,26.3 optical data, Hecht and Gmber (*) 110 13.6 magnetic susceptibility, Gruber and

Hecht (*)

75 9.3 magnetic susceptibility, Mulak and

~oinierek (*)

91 11.3 neutron spectroscopy [52]

(*) References given in [52].

UPd, [53]. The results are interpreted as CEF transi- tions from the ground level to the level at

-

^{3 meV}

(35 K) and to another excited level at

-

¹⁴^meV

(165 K). The authors were unable to observe a transition from the ground level to the level at

-

³meV. Moreover, it was impossible to specify the levels. However, it seems that above mentioned works give substantial information for our understanding of the crystal field problem in actinides. We hope that similar contribution has been provided by the theoretical works presenting the ligand field theory for actinides [54] and parametrization of crystal induced correlation between f electrons [55].

The aim of both papers is finding a physically justifiable way to reduce the number of the f electron correlation crystal field parameters below the number required according to simple phenomenolo- gical consideration.

(6)

THE STRUCTURE, MAGNETIC AND RELATED PROPERTIES

Ref ere

[I] TRoC, R., NIEDZIELSKI, R., LECIEJEWICZ, J., MURASIK, A., Rare Earths and Actinides, Durham 1977, Inst. Phys.

Conf. Proc. Series No 37 C4-7, p. 196.

[2] DAMIEN, D., BERGER, R., J. Inorg. Nucl. Chem., Supplement 1976, p. 109.

[3] ZACHARIASEN, W. H., Acta Cryst. 2 (1949) 57.

[4] CARTER, F. L., MILLER, R. C., RYAN, F. M., Adv. Energy Conversion 1 (1961) 165 ; CARTER, F. L., J. Solid State Chem. 5 (1972) 300.

151 SUSKI, W., JANUS, B., to be published.

[6] ALLBUTT, M., DELL, R. M., JUNKINSON, A. R., in The Chemistry o f Extended Defects in Non-Metallic Solids, (North Holland, Amsterdam) 1970, p. 124.

[7] MITCHELL, A. W., LAM, D. J., unpublished results.

[8] The Actinides. Electronic Structure and Related Properties, Eds. A. J. Freeman and J. B. Darby Jr. (Academic Press, New York) 1974.

[9] SUSKI, W., in Plutonium and Other Actinides, Eds. H. Blank and R. Lindner (North-Holland. Amsterdam) 1976.

p. 621.

r1Ol TROC. R.. MULAK, J., SUSKI. W., P h ~ s . Status Solidi 43

[11] PRZYSTAWA, J., J. Phys Chem. Solids 31 (1970) 2158 ; DOBRZYNSKI, L., PRZYSTAWA, J., Phys. Status Solidi 42 (1970) K15 ; PRZYSTAWA, J., PRAVECZKI, E., J. Phys.

Chem. Solids 33 (1972) 1943.

[12] BUHRER, C. F., J. Phys. Chem. Solids 30 (1969) 1273.

[13] MURASIK, A., LIGENZA, S., LECIEJEWICZ, J., TRoC, R., Phys.

Status solidi ( a ) 19 (1972) 1943.

[I41 WALASEK, K., PRZYSTAWA, J., PAWLIKOWSKI, J., J. Phys. C 9 (1976) 1277.

[15] FILKA, S., STERNBERK, J., MESJOVSKP, A., ZENTKO, A., Czech. J. Phys. B 27 (1977) 182.

[I61 NOVOTNP, P., STERNBERK, J., Phys. Status Solidi ( a ) 33 (1976) K15.

[17] KACZER, J., NOVOTNP, P., Phys. Status solidi ( a ) 39 (1977) 351.

[I81 NOVOTNP, P., ROSKOVEC, V., STERNBERK, J., Phys. Status Solidi ( b ) 83 (1977) K87.

[19] K O L A ~ E K , J., SEDLAK, B., M E ~ ~ O V S K Y , A., Czech. J. Phys.

B 27 (1977) 797.

[20] ALLES, A., FALK, B. G., WESTRUM, Jr. E. F., G R ~ N V O L D , F., ZAKI, M. R., J. Inorg. Nucl. Chem. 39 (1977) 1993.

[21] HENKIE, Z., KLAMUT, J., Physica 86-88B (1977) 991.

[22] KIM, D. J., Progr. Theor. Phys. 31 (1964) 921.

[23] HENKIE, Z., in the press.

[24] KLAMUT, J., MULAK, J., SUSKI, W., ROC. 2nd Intern. c o d . Electr. Str. Actin., eds. J. Mulak, W. Suski and R. TroC (Ossolineum, Wroclaw) 1977, p. 389.

[25] SIMA, V., STERNBERK, J., SMETANA, Z., SECHOVSKP, V., Czech. J. Phys. B 26 (1976) 1167.

[26] LECIEJEWICZ, J., private communication.

nces

[27] RAPHAEL, D., DENOVION, C.-H., J. Physique 30 (1969) 261.

[28] DUNLAP, B. D., LAM, D. J., KALVIUS, G. M., SHENOY, G . K . , J. Appl. Phys. 42 (1971) 1719.

1291 HENKIE, Z., MARKOWSKI, P. I., J. Cryst. Growth 41 (1978) 303.

[30] HENKIE, Z., MARKOWSKI, P. J., J. Phys Chem. Solids 39 (1978) 39.

[31] MARKOWSKI, P. J., HENKIE, Z., WOJAKOWSKI, A., in the press.

[32] ZYGMUNT, A., Phys. Status Solidi ( a ) 43 (1977) 573.

[33] PTASIEWICZ-BAK, H., LECIEJEWICZ, J., ZYGMUNT, A., Phys.

Status Solidi ( a ) 47 (1978) 349.

[34]. ~ ~ E N I E R E K , Z., n o t , R., J. Magn. Materials 8 (1978) 210.

[35] LECIEJEWICZ, J., ~%OENIEREK, Z., TROC, R., Solid State Commun. 22 (1977) 697.

[36] SCHINKEL, C. J., ZOENIEREK, Z., TRoC, R., Phys Status Solidi ( a ) 45 (1978) K39.

[37] ZYGMUNT, A., private communication.

[38] WOJAKOWSKI, A., HENKIE, Z., Acta Phys. Polon. A 52 (1977) 401.

[39] ZOWIEREK, Z., ROC. 3rd Intern. Conf. Electr. Str. Actin., Grenoble 1978.

[40] NOEL, H., TRoC, R., J. Solid State Chem., in the press.

[41] MULAK, J., J. Solid State Chem. 21 (1977) 117.

[42] MULAK, J., J. Solid State Chem. 25 (1978) 355.

[43] MULAK, J., ZOENIEREK, Z., Solid State Cornmun. 26 (1978) 275.

[44] ZOENIEREK, Z., Solid State Commun. 26 (1978) 307.

[45] JEZOWSKA-TRZEBIATOWSKA, B., JANCZAK, C. K., MULAK, J., Chem. Phys. Lett., in the press.

[46] PRIFFORD, M. M., SOULIB, E., Coll. Intern. CNRS, Spectro- scopie des ions de transition et des ClCments lourds dans les solides, Lyon, France, 28 June-3 July 1976, CEA.

Conf. 3679.

[47] MIYAKE, C., FUJI, K., IMOTO, S., Inorg. Nucl. Chem. Lett.

13 (1977) 53.

[48] MIYAKE, C., FUJI, K., IMOTO, S., Chem. Phys. Lett. 46 (1977) 349.

[49] LYON, W. G., OSBORNE, D. W., FLOTOW, H. E., HOEKSTRA, H. R., J. Chem. Thermodynamics 9 (1977) 201.

[50] LANDER, G. H., in Crystal Field Effects in Metals and Alloys, ed. A. Furer (Plenum Press, New York) 1977, p. 213.

[51] FURRER, A., MURASIK, A., VOGT, O., Helv. Physica Acta 50 (1977) 447.

[52] MURASIK, A,, LECIEJEWICZ, J., ZOENIEREK, Z., Phys Status Solidi (b) 80 (1977) K137.

[53] SHAMIR, N., MELAMUD, M., SHAKED, H., WEGER, M., Meeting in Israel Physical Society (1977) 62-63.

[54] JUDD, B. R., J. Chem. Phys. 66 (1977) 3163.

[55] NEWMAN, D. J., J. Phys. C 10 (1977) 4753.

(7)

W. SUSKI

DISCUSSION Dr. G. H. LANDER.

-

In the U,Te, do you have

single crystals, because without them magnetization measurements are very difficult to interpret ?

W. SUSKI. - We have not single crystals of the U,Te,, thus we do not try to make detailed interpretation. For the time-being we just present the transition temperatures which seem to be reasonable determined from powder data.

Dr. ROSSAT-MIGNOD.

-

U,As, has been investigated by neutron diffraction at the CEN-Grenoble.

On single crystal, small superlattice reflexions (1 10) and (200) have been detected with unpolarized neutrons. Unfortunately very strong extinction effects do not allow a quantitative treatment of the data. In addition, a polarized neutron experiment has been performed and it seems that the magnetic structure of U,As, could be described by a non-collinear arrangement of moments tilted from the / I l l / axis Dr. J. GAL. - Did you try to go to higher applied

by only a low angle ( = 10"). This result is an fields ? It may show that the compound (U2Te,,,) is

agreement with prediction of Przystawa [I] based on ferromagnetic and what one see is alignment of the

symmetry considerations. It leads to an ordered magnetic domains ?

moment lower than those derived from the Curie- W. SUSK[.

-

~t could be but up to 50 k o e we have Weiss law in agreement with magnetization data [2].

observed only shift of temperature but no change of [l] PRZYSTAWA, J., J. ~ h y s . Chem. Solids 31 (1970) 2158 ; 33

the curve character. (1972) 1943.

[2] BUHRER, C. F., J. Phys. Chem. Solids 30 (1969) 1272.

Dr. KOELLING. - Is the Th,P, structure unique to

rare earths and actinides ? W. SUSKI.

-

I knew about the complexity of your results but because I learn a little bit more about an W. SUSKI.

-

According to my best knowledge : interpretation only today, I have not included these

Yes. informations in my talk.

COMMENT

Pr. J . LECIEJEWICZ. - Neutron diffraction mea- tic unit cells are multiples ( n = 3 and 4) of the

surements on polycrystalline samples of UPd2Ge2 chemical ones in the direction of the c-axis. The and UPd,Si2 (reported by us during the XI congress alignment of moments is sinusoidally modulated. We of crystallography in Warszaw) have shown that did not find any evidence for ferromagnetic order in both compounds exhibit antiferromagnetic ordering both compounds down to 4 K.

below the respective N6el temperatures. The magne-